Skip to contentSkip to conversation

drm/amd: Add USBC connector ID

PATCH New Sasha Levin 1 participant 11 messages Apr 12, 2022

Subsystems (8)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/11 drm/amd: Add USBC connector ID

From: Aurabindo Pillai <[email protected]>

[ Upstream commit c5c948aa894a831f96fccd025e47186b1ee41615 ]

[Why&How] Add a dedicated AMDGPU specific ID for use with newer ASICs that support USB-C output

---
 drivers/gpu/drm/amd/amdgpu/ObjectID.h | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/amd/amdgpu/ObjectID.h +1 -0
index 06192698bd96..c90567de8bf7 100644
@@ -119,6 +119,7 @@
 #define CONNECTOR_OBJECT_ID_eDP                   0x14
 #define CONNECTOR_OBJECT_ID_MXM                   0x15
 #define CONNECTOR_OBJECT_ID_LVDS_eDP              0x16
+#define CONNECTOR_OBJECT_ID_USBC                  0x17
 
 /* deleted */
 
Signature · 1 line
2.35.1
author maintainer
#1 lore raw

PATCH 2/11 drm/amdkfd: Check for potential null return of kmalloc_array()

From: QintaoShen <[email protected]>

[ Upstream commit ebbb7bb9e80305820dc2328a371c1b35679f2667 ]

As the kmalloc_array() may return null, the 'event_waiters[i].wait' would lead to null-pointer dereference. Therefore, it is better to check the return value of kmalloc_array() to avoid this confusion.

---
 drivers/gpu/drm/amd/amdkfd/kfd_events.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdkfd/kfd_events.c +2 -0
index 944abfad39c1..1d8dd81dfc70 100644
@@ -607,6 +607,8 @@ static struct kfd_event_waiter *alloc_event_waiters(uint32_t num_events)
 	event_waiters = kmalloc_array(num_events,
 					sizeof(struct kfd_event_waiter),
 					GFP_KERNEL);
+	if (!event_waiters)
+		return NULL;
 
 	for (i = 0; (event_waiters) && (i < num_events) ; i++) {
 		INIT_LIST_HEAD(&event_waiters[i].waiters);
Signature · 1 line
2.35.1
author maintainer
#7 lore raw

PATCH 8/11 gpu: ipu-v3: Fix dev_dbg frequency output

In reply to Sasha Levin

From: Leo Ruan <[email protected]>

[ Upstream commit 070a88fd4a03f921b73a2059e97d55faaa447dab ]

This commit corrects the printing of the IPU clock error percentage if it is between -0.1% to -0.9%. For example, if the pixel clock requested is 27.2 MHz but only 27.0 MHz can be achieved the deviation is -0.8%. But the fixed point math had a flaw and calculated error of 0.2%.

Before:
Clocks: IPU 270000000Hz DI 24716667Hz Needed 27200000Hz IPU clock can give 27000000 with divider 10, error 0.2% Want 27200000Hz IPU 270000000Hz DI 24716667Hz using IPU, 27000000Hz

After:
Clocks: IPU 270000000Hz DI 24716667Hz Needed 27200000Hz IPU clock can give 27000000 with divider 10, error -0.8% Want 27200000Hz IPU 270000000Hz DI 24716667Hz using IPU, 27000000Hz

---
 drivers/gpu/ipu-v3/ipu-di.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/gpu/ipu-v3/ipu-di.c +3 -2
index d2f1bd9d3deb..c498dc7d8838 100644
@@ -460,8 +460,9 @@ static void ipu_di_config_clock(struct ipu_di *di,
 
 		error = rate / (sig->mode.pixelclock / 1000);
 
-		dev_dbg(di->ipu->dev, "  IPU clock can give %lu with divider %u, error %d.%u%%\n",
-			rate, div, (signed)(error - 1000) / 10, error % 10);
+		dev_dbg(di->ipu->dev, "  IPU clock can give %lu with divider %u, error %c%d.%d%%\n",
+			rate, div, error < 1000 ? '-' : '+',
+			abs(error - 1000) / 10, abs(error - 1000) % 10);
 
 		/* Allow a 1% error */
 		if (error < 1010 && error >= 990) {
Signature · 1 line
2.35.1
author maintainer
#8 lore raw

PATCH 9/11 arm64: alternatives: mark patch_alternative() as `noinstr`

In reply to Sasha Levin

From: Joey Gouly <[email protected]>

[ Upstream commit a2c0b0fbe01419f8f5d1c0b9c581631f34ffce8b ]

The alternatives code must be noinstr such that it does not patch itself, as the cache invalidation is only performed after all the alternatives have been applied.

Mark patch_alternative() as noinstr. Mark branch_insn_requires_update() and get_alt_insn() with __always_inline since they are both only called through patch_alternative().

Booting a kernel in QEMU TCG with KCSAN=y and ARM64_USE_LSE_ATOMICS=y caused a boot hang:

[    0.241121] CPU: All CPU(s) started at EL2

The alternatives code was patching the atomics in __tsan_read4() from LL/SC atomics to LSE atomics.

The following fragment is using LL/SC atomics in the .text section:

  | <__tsan_unaligned_read4+304>:     ldxr    x6, [x2]
  | <__tsan_unaligned_read4+308>:     add     x6, x6, x5
  | <__tsan_unaligned_read4+312>:     stxr    w7, x6, [x2]
  | <__tsan_unaligned_read4+316>:     cbnz    w7, <__tsan_unaligned_read4+304>

This LL/SC atomic sequence was to be replaced with LSE atomics. However since the alternatives code was instrumentable, __tsan_read4() was being called after only the first instruction was replaced, which led to the following code in memory:

  | <__tsan_unaligned_read4+304>:     ldadd   x5, x6, [x2]
  | <__tsan_unaligned_read4+308>:     add     x6, x6, x5
  | <__tsan_unaligned_read4+312>:     stxr    w7, x6, [x2]
  | <__tsan_unaligned_read4+316>:     cbnz    w7, <__tsan_unaligned_read4+304>

This caused an infinite loop as the stxr instruction never completed successfully, so w7 was always 0.

---
 arch/arm64/kernel/alternative.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/arm64/kernel/alternative.c +3 -3
index 4c385763c361..1398fd816068 100644
@@ -42,7 +42,7 @@ struct alt_region {
 /*
  * Check if the target PC is within an alternative block.
  */
-static bool branch_insn_requires_update(struct alt_instr *alt, unsigned long pc)
+static __always_inline bool branch_insn_requires_update(struct alt_instr *alt, unsigned long pc)
 {
 	unsigned long replptr = (unsigned long)ALT_REPL_PTR(alt);
 	return !(pc >= replptr && pc <= (replptr + alt->alt_len));
@@ -50,7 +50,7 @@ static bool branch_insn_requires_update(struct alt_instr *alt, unsigned long pc)
 
 #define align_down(x, a)	((unsigned long)(x) & ~(((unsigned long)(a)) - 1))
 
-static u32 get_alt_insn(struct alt_instr *alt, __le32 *insnptr, __le32 *altinsnptr)
+static __always_inline u32 get_alt_insn(struct alt_instr *alt, __le32 *insnptr, __le32 *altinsnptr)
 {
 	u32 insn;
 
@@ -95,7 +95,7 @@ static u32 get_alt_insn(struct alt_instr *alt, __le32 *insnptr, __le32 *altinsnp
 	return insn;
 }
 
-static void patch_alternative(struct alt_instr *alt,
+static noinstr void patch_alternative(struct alt_instr *alt,
 			      __le32 *origptr, __le32 *updptr, int nr_inst)
 {
 	__le32 *replptr;
Signature · 7 lines
2.35.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#4 lore raw

PATCH 5/11 scsi: ibmvscsis: Increase INITIAL_SRP_LIMIT to 1024

In reply to Sasha Levin

From: Tyrel Datwyler <[email protected]>

[ Upstream commit 0bade8e53279157c7cc9dd95d573b7e82223d78a ]

The adapter request_limit is hardcoded to be INITIAL_SRP_LIMIT which is currently an arbitrary value of 800. Increase this value to 1024 which better matches the characteristics of the typical IBMi Initiator that supports 32 LUNs and a queue depth of 32.

This change also has the secondary benefit of being a power of two as required by the kfifo API. Since, Commit ab9bb6318b09 ("Partially revert "kfifo: fix kfifo_alloc() and kfifo_init()"") the size of IU pool for each target has been rounded down to 512 when attempting to kfifo_init() those pools with the current request_limit size of 800.

---
 drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c +1 -1
index 25d2741cdf96..b97140661119 100644
@@ -44,7 +44,7 @@
 
 #define IBMVSCSIS_VERSION	"v0.2"
 
-#define	INITIAL_SRP_LIMIT	800
+#define	INITIAL_SRP_LIMIT	1024
 #define	DEFAULT_MAX_SECTORS	256
 #define MAX_TXU			1024 * 1024
 
Signature · 1 line
2.35.1
author maintainer
#10 lore raw

PATCH 11/11 drivers: net: slip: fix NPD bug in sl_tx_timeout()

In reply to Sasha Levin

From: Duoming Zhou <[email protected]>

[ Upstream commit ec4eb8a86ade4d22633e1da2a7d85a846b7d1798 ]

When a slip driver is detaching, the slip_close() will act to cleanup necessary resources and sl->tty is set to NULL in slip_close(). Meanwhile, the packet we transmit is blocked, sl_tx_timeout() will be called. Although slip_close() and sl_tx_timeout() use sl->lock to synchronize, we don`t judge whether sl->tty equals to NULL in sl_tx_timeout() and the null pointer dereference bug will happen.

   (Thread 1)                 |      (Thread 2)
                              | slip_close()
                              |   spin_lock_bh(&sl->lock)
                              |   ...
...                           |   sl->tty = NULL //(1)
sl_tx_timeout()               |   spin_unlock_bh(&sl->lock)
  spin_lock(&sl->lock);       |
  ...                         |   ...
  tty_chars_in_buffer(sl->tty)|
    if (tty->ops->..) //(2)   |
    ...                       |   synchronize_rcu()

We set NULL to sl->tty in position (1) and dereference sl->tty in position (2).

This patch adds check in sl_tx_timeout(). If sl->tty equals to NULL, sl_tx_timeout() will goto out.

---
 drivers/net/slip/slip.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/slip/slip.c +1 -1
index d7882b548b79..dd95d6181afa 100644
@@ -471,7 +471,7 @@ static void sl_tx_timeout(struct net_device *dev)
 	spin_lock(&sl->lock);
 
 	if (netif_queue_stopped(dev)) {
-		if (!netif_running(dev))
+		if (!netif_running(dev) || !sl->tty)
 			goto out;
 
 		/* May be we must check transmitter timeout here ?
Signature · 1 line
2.35.1
author maintainer
#3 lore raw

PATCH 4/11 scsi: target: tcmu: Fix possible page UAF

In reply to Sasha Levin

From: Xiaoguang Wang <[email protected]>

[ Upstream commit a6968f7a367f128d120447360734344d5a3d5336 ]

tcmu_try_get_data_page() looks up pages under cmdr_lock, but it does not take refcount properly and just returns page pointer. When tcmu_try_get_data_page() returns, the returned page may have been freed by tcmu_blocks_release().

We need to get_page() under cmdr_lock to avoid concurrent tcmu_blocks_release().

---
 drivers/target/target_core_user.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/target/target_core_user.c +2 -1
index 96601fda47b1..f687481ccfdc 100644
@@ -1165,6 +1165,7 @@ static struct page *tcmu_try_get_block_page(struct tcmu_dev *udev, uint32_t dbi)
 	mutex_lock(&udev->cmdr_lock);
 	page = tcmu_get_block_page(udev, dbi);
 	if (likely(page)) {
+		get_page(page);
 		mutex_unlock(&udev->cmdr_lock);
 		return page;
 	}
@@ -1233,6 +1234,7 @@ static int tcmu_vma_fault(struct vm_fault *vmf)
 		/* For the vmalloc()ed cmd area pages */
 		addr = (void *)(unsigned long)info->mem[mi].addr + offset;
 		page = vmalloc_to_page(addr);
+		get_page(page);
 	} else {
 		uint32_t dbi;
 
@@ -1243,7 +1245,6 @@ static int tcmu_vma_fault(struct vm_fault *vmf)
 			return VM_FAULT_NOPAGE;
 	}
 
-	get_page(page);
 	vmf->page = page;
 	return 0;
 }
Signature · 1 line
2.35.1
author maintainer
#2 lore raw

PATCH 3/11 Drivers: hv: vmbus: Prevent load re-ordering when reading ring buffer

In reply to Sasha Levin

From: Michael Kelley <[email protected]>

[ Upstream commit b6cae15b5710c8097aad26a2e5e752c323ee5348 ]

When reading a packet from a host-to-guest ring buffer, there is no memory barrier between reading the write index (to see if there is a packet to read) and reading the contents of the packet. The Hyper-V host uses store-release when updating the write index to ensure that writes of the packet data are completed first. On the guest side, the processor can reorder and read the packet data before the write index, and sometimes get stale packet data. Getting such stale packet data has been observed in a reproducible case in a VM on ARM64.

Fix this by using virt_load_acquire() to read the write index, ensuring that reads of the packet data cannot be reordered before it. Preventing such reordering is logically correct, and with this change, getting stale data can no longer be reproduced.

---
 drivers/hv/ring_buffer.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/hv/ring_buffer.c +10 -1
index 74c1dfb8183b..6b08e9d9b382 100644
@@ -340,7 +340,16 @@ int hv_ringbuffer_read(struct vmbus_channel *channel,
 static u32 hv_pkt_iter_avail(const struct hv_ring_buffer_info *rbi)
 {
 	u32 priv_read_loc = rbi->priv_read_index;
-	u32 write_loc = READ_ONCE(rbi->ring_buffer->write_index);
+	u32 write_loc;
+
+	/*
+	 * The Hyper-V host writes the packet data, then uses
+	 * store_release() to update the write_index.  Use load_acquire()
+	 * here to prevent loads of the packet data from being re-ordered
+	 * before the read of the write_index and potentially getting
+	 * stale data.
+	 */
+	write_loc = virt_load_acquire(&rbi->ring_buffer->write_index);
 
 	if (write_loc >= priv_read_loc)
 		return write_loc - priv_read_loc;
Signature · 1 line
2.35.1
author maintainer
#5 lore raw

PATCH 6/11 net: micrel: fix KS8851_MLL Kconfig

In reply to Sasha Levin

From: Randy Dunlap <[email protected]>

[ Upstream commit c3efcedd272aa6dd5929e20cf902a52ddaa1197a ]

KS8851_MLL selects MICREL_PHY, which depends on PTP_1588_CLOCK_OPTIONAL, so make KS8851_MLL also depend on PTP_1588_CLOCK_OPTIONAL since 'select' does not follow any dependency chains.

Fixes kconfig warning and build errors:

WARNING: unmet direct dependencies detected for MICREL_PHY
Depends on [m]: NETDEVICES [=y] && PHYLIB [=y] && PTP_1588_CLOCK_OPTIONAL [=m] Selected by [y]:
- KS8851_MLL [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_MICREL [=y] && HAS_IOMEM [=y]

ld: drivers/net/phy/micrel.o: in function `lan8814_ts_info':
micrel.c:(.text+0xb35): undefined reference to `ptp_clock_index'
ld: drivers/net/phy/micrel.o: in function `lan8814_probe':
micrel.c:(.text+0x2586): undefined reference to `ptp_clock_register'

---
 drivers/net/ethernet/micrel/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/micrel/Kconfig +1 -0
index b7e2f49696b7..aa12bace8673 100644
@@ -45,6 +45,7 @@ config KS8851
 config KS8851_MLL
 	tristate "Micrel KS8851 MLL"
 	depends on HAS_IOMEM
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select MII
 	---help---
 	  This platform driver is for Micrel KS8851 Address/data bus
Signature · 1 line
2.35.1
author maintainer
#6 lore raw

PATCH 7/11 ata: libata-core: Disable READ LOG DMA EXT for Samsung 840 EVOs

In reply to Sasha Levin

From: Christian Lamparter <[email protected]>

[ Upstream commit 5399752299396a3c9df6617f4b3c907d7aa4ded8 ]

Samsung' 840 EVO with the latest firmware (EXT0DB6Q) locks up with the a message: "READ LOG DMA EXT failed, trying PIO" during boot.

Initially this was discovered because it caused a crash with the sata_dwc_460ex controller on a WD MyBook Live DUO.

The reporter "Tice Rex" which has the unique opportunity that he has two Samsung 840 EVO SSD! One with the older firmware "EXT0BB0Q" which booted fine and didn't expose "READ LOG DMA EXT". But the newer/latest firmware "EXT0DB6Q" caused the headaches.

---
 drivers/ata/libata-core.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/ata/libata-core.c +3 -0
index 791374199e22..d3a7b3bb5043 100644
@@ -4588,6 +4588,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 						ATA_HORKAGE_ZERO_AFTER_TRIM, },
 	{ "Crucial_CT*MX100*",		"MU01",	ATA_HORKAGE_NO_NCQ_TRIM |
 						ATA_HORKAGE_ZERO_AFTER_TRIM, },
+	{ "Samsung SSD 840 EVO*",	NULL,	ATA_HORKAGE_NO_NCQ_TRIM |
+						ATA_HORKAGE_NO_DMA_LOG |
+						ATA_HORKAGE_ZERO_AFTER_TRIM, },
 	{ "Samsung SSD 840*",		NULL,	ATA_HORKAGE_NO_NCQ_TRIM |
 						ATA_HORKAGE_ZERO_AFTER_TRIM, },
 	{ "Samsung SSD 850*",		NULL,	ATA_HORKAGE_NO_NCQ_TRIM |
Signature · 1 line
2.35.1
author maintainer
#9 lore raw

PATCH 10/11 scsi: mvsas: Add PCI ID of RocketRaid 2640

In reply to Sasha Levin

From: Alexey Galakhov <[email protected]>

[ Upstream commit 5f2bce1e222028dc1c15f130109a17aa654ae6e8 ]

The HighPoint RocketRaid 2640 is a low-cost SAS controller based on Marvell chip. The chip in question was already supported by the kernel, just the PCI ID of this particular board was missing.

---
 drivers/scsi/mvsas/mv_init.c | 1 +
 1 file changed, 1 insertion(+)
drivers/scsi/mvsas/mv_init.c +1 -0
index 718c88de328b..eada13c775ba 100644
@@ -678,6 +678,7 @@ static struct pci_device_id mvs_pci_table[] = {
 	{ PCI_VDEVICE(ARECA, PCI_DEVICE_ID_ARECA_1300), chip_1300 },
 	{ PCI_VDEVICE(ARECA, PCI_DEVICE_ID_ARECA_1320), chip_1320 },
 	{ PCI_VDEVICE(ADAPTEC2, 0x0450), chip_6440 },
+	{ PCI_VDEVICE(TTI, 0x2640), chip_6440 },
 	{ PCI_VDEVICE(TTI, 0x2710), chip_9480 },
 	{ PCI_VDEVICE(TTI, 0x2720), chip_9480 },
 	{ PCI_VDEVICE(TTI, 0x2721), chip_9480 },
Signature · 1 line
2.35.1